]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/vine.py
Prepare for upload.
[youtubedl] / youtube_dl / extractor / vine.py
index b44b1cb024f607e78ba1b977be42220ea7090e81..42995226e584b0ce4e0d207b6b702157ec6f4030 100644 (file)
@@ -1,37 +1,98 @@
+from __future__ import unicode_literals
+
 import re
+import json
+import itertools
 
 from .common import InfoExtractor
+from ..utils import unified_strdate
 
 
 class VineIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?vine\.co/v/(?P<id>\w+)'
+    _VALID_URL = r'https?://(?:www\.)?vine\.co/v/(?P<id>\w+)'
+    _TEST = {
+        'url': 'https://vine.co/v/b9KOOWX7HUx',
+        'md5': '2f36fed6235b16da96ce9b4dc890940d',
+        'info_dict': {
+            'id': 'b9KOOWX7HUx',
+            'ext': 'mp4',
+            'title': 'Chicken.',
+            'description': 'Chicken.',
+            'upload_date': '20130519',
+            'uploader': 'Jack Dorsey',
+            'uploader_id': '76',
+        },
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-
         video_id = mobj.group('id')
-        webpage_url = 'https://vine.co/v/' + video_id
-        webpage = self._download_webpage(webpage_url, video_id)
 
-        self.report_extraction(video_id)
+        webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id)
+
+        data = json.loads(self._html_search_regex(
+            r'window\.POST_DATA = { %s: ({.+?}) }' % video_id, webpage, 'vine data'))
+
+        formats = [
+            {
+                'url': data['videoLowURL'],
+                'ext': 'mp4',
+                'format_id': 'low',
+            },
+            {
+                'url': data['videoUrl'],
+                'ext': 'mp4',
+                'format_id': 'standard',
+            }
+        ]
 
-        video_url = self._html_search_regex(r'<meta property="twitter:player:stream" content="(.+?)"',
-            webpage, u'video URL')
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'description': data['description'],
+            'thumbnail': data['thumbnailUrl'],
+            'upload_date': unified_strdate(data['created']),
+            'uploader': data['username'],
+            'uploader_id': data['userIdStr'],
+            'like_count': data['likes']['count'],
+            'comment_count': data['comments']['count'],
+            'repost_count': data['reposts']['count'],
+            'formats': formats,
+        }
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
 
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)(\?.*?)?"',
-            webpage, u'thumbnail', fatal=False)
+class VineUserIE(InfoExtractor):
+    IE_NAME = 'vine:user'
+    _VALID_URL = r'(?:https?://)?vine\.co/(?P<user>[^/]+)/?(\?.*)?$'
+    _VINE_BASE_URL = "https://vine.co/"
+    _TEST = {
+        'url': 'https://vine.co/Visa',
+        'info_dict': {
+            'id': 'Visa',
+        },
+        'playlist_mincount': 46,
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        user = mobj.group('user')
+
+        profile_url = "%sapi/users/profiles/vanity/%s" % (
+            self._VINE_BASE_URL, user)
+        profile_data = self._download_json(
+            profile_url, user, note='Downloading user profile data')
 
-        uploader = self._html_search_regex(r'<div class="user">.*?<h2>(.+?)</h2>',
-            webpage, u'uploader', fatal=False, flags=re.DOTALL)
+        user_id = profile_data['data']['userId']
+        timeline_data = []
+        for pagenum in itertools.count(1):
+            timeline_url = "%sapi/timelines/users/%s?page=%s" % (
+                self._VINE_BASE_URL, user_id, pagenum)
+            timeline_page = self._download_json(
+                timeline_url, user, note='Downloading page %d' % pagenum)
+            timeline_data.extend(timeline_page['data']['records'])
+            if timeline_page['data']['nextPage'] is None:
+                break
 
-        return [{
-            'id':        video_id,
-            'url':       video_url,
-            'ext':       'mp4',
-            'title':     video_title,
-            'thumbnail': thumbnail,
-            'uploader':  uploader,
-        }]
+        entries = [
+            self.url_result(e['permalinkUrl'], 'Vine') for e in timeline_data]
+        return self.playlist_result(entries, user)